New calculations of the saturnian He 584 Angstrom airglow intensity, u
sing radiative transfer models with partial frequency redistribution a
nd inhomogeneous atmospheric models, are presented. Our reference cond
itions assume a helium mixing ratio, f(Hc), Of 0.033 and a He 584 Angs
trom solar Aux of 2 x 10(9) photons cm(-2) s(-1) at 1 AU, typical of t
he EUVAC model, with a solar line FWHM of 120 m Angstrom and an atmosp
here consistent with the Voyager UVS occultation results. For these co
nditions we require the eddy diffusion coefficients at the homopause,
K-h, to be greater than 10(9) cm(2) s(-1) in order to fit the Ultravio
let Spectrometer measurements of Voyager 1 and 2 He 584 Angstrom airgl
ow measurements. These values of K-h are even higher than obtained in
the earlier work of Sandel ef al. (1982, Geophys. Res, Lett, 9, 1077-1
080) and Atreya (1982, Planet. Space Sci. 30, 849-854) and seem unreas
onably high. This suggests that either the values of one or more of th
e parameters of our model are not correct or the measured UVS airglow
is too bright and that there is a problem with calibration. Even so, w
e suggest that K-h is likely to be greater than 10(8) cm(2) s(-1) duri
ng the period of the Voyager encounters. Only if we increase the heliu
m mixing ratio to the solar value and use the EUVT94 solar flux model
for He 584 Angstrom is it possible to begin to reconcile K-h, from the
He 584 Angstrom airglow and the work of Smith er al, (1983, J. Geophy
s. Res. 88, 8667-8678), (C) 1998 Academic Press.